Downstaging the most deadly cancer is cost-effective. Can we achieve a cost-effective early diagnosis?

Downstaging the most deadly cancer is cost-effective. Can we achieve a cost-effective early diagnosis?

Background: Early diagnosis of pancreatic cancer, with less than 5 years for almost all patients --even if diagnosed in resectable stage with curative intent rection— has been analysed for cost-effectiveness in some UK studies. The existing cost-effectiveness studies rely on the accuracy of imaging diagnostics to guide treatment, including avoiding unnecessary resections.

Aim: An effect of early diagnosis is downstaging which shifts the cohort of patients entering treatment with more patients in resectable and less patients in metastatic stages. The aim of this study is to estimate an acceptable range for a diagnostic cost by assuming that the unique effect of early diagnosis is downstaging as the unique treatment effect that results in health gains (measured in quality-adjusted life years, QALYs) and incremental costs.

Methods: We propose a two-part method for diagnostic cost-effectiveness evaluation to establish a value -based pricing that uniquely considers value from downstaging. The first part of the model is the last part of a Markov model decision tree simulating two identical treatment pathways in survival and costs, except by the initial cohort, with the early diagnosis downstaging cohort effect assumed as in Sweden trial for screening diabetic patients-at-risk of pancreatic cancer (Ghatnekar et al., 2013). According to the incremental cost-effectiveness ratio (ICER) estimated in the first part, we set at a second “residual cost-effectiveness threshold” at £30,000 per QALY minus ICER, considering the £30K NICE threshold. The acceptable price for the diagnostic is then calculated backwards in the decision tree according to prevalence and accuracy parameters.

Results and conclusions: The average ICER of the first part of downstaging averages around £19,000 per QALY across the results of the probabilistic sensitivity analysis (PSA) with 1,000 random draws allowing large degree of uncertainty for all parameters, including survival, costs, and utilities. Therefore, the introduction of a diagnostic could be cost effective of the second part with an ICER in the decision tree below £11,000 per QALY. Accounting for an at-risk population with a prevalence of 8% of pancreatic patients, there is still a positive acceptable incremental cost around £37-£55 for the early diagnostic.